Calculator Models Hardware Compare

Product

Calculator
Compare
Tier List

Browse

Models
Hardware
Docs

About

Why It Works
What's New
Legal Notice
Privacy Policy

All estimates are approximations based on mathematical models and public specifications. Actual performance may vary. Do not make purchasing decisions based solely on these estimates.

Data sourced from Hugging Face, Ollama, and official model documentation. Model names and logos are trademarks of their respective owners.

© 2026 Will It Run AI — Fase Consulting Ibiza, S.L. (NIF: B57969656)

Home/Hardware/GPUs/Quadro RTX 8000 48GB

NVIDIA

Quadro RTX 8000 48GB

Quadro RTXWorkstationTuringPCIe 3CUDA

48GB

VRAM

672GB/s

Bandwidth

32TFLOPS

FP16 Compute

256TOPS

INT8 Inference

Quadro RTX 8000 48GBCategory AvgNVIDIA A16 64GB

Specifications

Compute

FP1632 TFLOPS

INT8256 TOPS

ArchitectureTuring

Memory

VRAM48 GB

Bandwidth672 GB/s

General

FamilyQuadro RTX

SegmentWorkstation

InterconnectPCIe 3

Compute PlatformCUDA

Architecture

Turing

Turing is NVIDIA's first-generation RTX architecture, introducing dedicated RT and Tensor Cores to consumer GPUs for the first time. Built on TSMC's 12nm FinFET process.

AI Relevance

The first consumer architecture with Tensor Cores, enabling meaningful acceleration for INT8 and FP16 inference. However, limited VRAM (typically 6-11 GB) restricts modern LLM model sizes.

Process: TSMC 12nmPlatform: CUDATensor Cores: Gen 2Precisions: FP32, FP16, INT8, INT4

Recommendations by Workload

Agentic Coding

C

Devstral Small 2 24B Instruct

This model is still usable for agentic-coding, but it is not the most specialized pick. It belongs to a current frontier family for local AI. It fits natively with comfortable headroom. Known channels: huggingface, ollama, lm-studio.

Decode 31.7 tok/s · 55K ctx · llama.cpp

27.8 GB / 48.0 GB VRAM

Chat

C

Qwen 3 32B

This model is a direct match for chat. It belongs to a current frontier family for local AI. It fits natively with comfortable headroom. Known channels: huggingface, ollama, lm-studio.

Decode 23.8 tok/s · 14K ctx · llama.cpp

27.7 GB / 48.0 GB VRAM

Coding

C

Devstral Small 2 24B Instruct

This model is a direct match for coding. It belongs to a current frontier family for local AI. It fits natively with comfortable headroom. Known channels: huggingface, ollama, lm-studio.

Decode 31.7 tok/s · 32K ctx · llama.cpp

24.1 GB / 48.0 GB VRAM

RAG

C

Command R 35B

This model is a direct match for rag. It sits in the middle of the current model mix. It fits natively with comfortable headroom. Known channels: huggingface, ollama, lm-studio.

Decode 21.7 tok/s · 40K ctx · llama.cpp

38.0 GB / 48.0 GB VRAM

Reasoning

C

Qwen 3 32B

This model is a direct match for reasoning. It belongs to a current frontier family for local AI. It fits natively with comfortable headroom. Known channels: huggingface, ollama, lm-studio.

Decode 23.8 tok/s · 25K ctx · llama.cpp

30.2 GB / 48.0 GB VRAM

Full Model Compatibility

Qwen3-Coder 30B A3B Instruct

30.5B25.1 GB65 tok/s31K ctx

Qwen3-VL 30B A3B Instruct

30B24.8 GB67 tok/s31K ctx

Qwen3.5 35B A3B

35B32.5 GB22 tok/s24K ctx

Qwen3.5 35B A3B

35B32.5 GB22 tok/s24K ctx

Qwen 2.5 Coder 32B

32B30.2 GB24 tok/s25K ctx

27B26.4 GB28 tok/s29K ctx

27B26.4 GB28 tok/s29K ctx

Devstral Small 2 24B Instruct

24B24.1 GB32 tok/s32K ctx

Devstral Small 1.1

24B24.1 GB32 tok/s32K ctx

Codestral 2 25.08

22B22.6 GB35 tok/s34K ctx

9B12.6 GB85 tok/s61K ctx

Qwen3.5 9B Uncensored HauhauCS Aggressive

9B12.6 GB85 tok/s61K ctx

9B12.6 GB85 tok/s61K ctx

Meta Llama 3.1 8B Instruct

8B11.8 GB95 tok/s65K ctx

llava llama 3 8b v1 1

8B11.8 GB95 tok/s65K ctx

DeepSeek R1 0528 Qwen3 8B

8B11.8 GB95 tok/s65K ctx

Meta Llama 3 8B Instruct

8B11.8 GB95 tok/s65K ctx

Llama 2 7B Chat

7B11.1 GB109 tok/s69K ctx

Mistral 7B Instruct v0.2

7B11.1 GB109 tok/s69K ctx

Mistral 7B Instruct v0.3

7B11.1 GB109 tok/s69K ctx

4B8.9 GB190 tok/s86K ctx

Llama 3.2 3B Instruct

3B8.7 GB219 tok/s89K ctx

4B8.9 GB190 tok/s86K ctx

Qwen2.5 3B Instruct

3B8.3 GB253 tok/s92K ctx

2B8.1 GB297 tok/s94K ctx

2B7.7 GB380 tok/s99K ctx

Gemmasutra Mini 2B v1

2B7.7 GB380 tok/s99K ctx

Qwen2.5 1.5B Instruct

1.5B7.4 GB464 tok/s104K ctx

Llama 3.2 1B Instruct Q8 0

1B7.3 GB487 tok/s105K ctx

TinyLlama 1.1B Chat v1.0

1.1B7.2 GB464 tok/s107K ctx

SmolVLM 500M Instruct

0.5B6.9 GB487 tok/s111K ctx

embeddinggemma 300M

0.3B6.7 GB487 tok/s114K ctx

Qwen3-Coder-Next

80B55.3 GB25 tok/s14K ctx

DeepSeek R1 671B

671B420.8 GB3 tok/s4K ctx

Devstral 2 123B Instruct

123B99.9 GB6 tok/s8K ctx

744B465.8 GB3 tok/s4K ctx

1000B620.7 GB2 tok/s4K ctx

Mistral Large 3

675B423.9 GB3 tok/s4K ctx

Mistral Small 4 119B

119B79.3 GB19 tok/s10K ctx

Qwen3-Coder 480B A35B Instruct

480B304.0 GB4 tok/s4K ctx

Qwen3.5 122B A10B

122B84.5 GB7 tok/s9K ctx

DeepSeek V3 671B

671B420.8 GB3 tok/s4K ctx

141B97.8 GB10 tok/s8K ctx

72B60.9 GB11 tok/s13K ctx

Qwen 3 235B A22B

235B152.5 GB9 tok/s5K ctx

Qwen3.5 397B A17B

397B309.9 GB2 tok/s4K ctx

70B59.3 GB11 tok/s13K ctx

Llama 4 Maverick 17B 128E

400B252.4 GB6 tok/s4K ctx

111B90.8 GB7 tok/s8K ctx

Qwen 2.5 VL 72B

72B60.9 GB11 tok/s13K ctx

Just out of reach

Models you could run with an upgrade

High-quality models that need a bit more memory

DeepSeek R1 671B

671BTier 5Needs ~426.6 GB

Devstral 2 123B Instruct

123BTier 5Needs ~119.2 GB

Runs on Mac Studio M3 Ultra 256GB

744BTier 5Needs ~472.0 GB

1000BTier 5Needs ~625.7 GB

Mistral Large 3

675BTier 5Needs ~430.3 GB

Upgrade paths

Upgrade from Quadro RTX 8000 48GB

See what you unlock with more powerful hardware

Upgrade options

Upgrade options

NVIDIA A16 64GBNext step up

64 GB VRAM (+16)

Unlocks Qwen 2.5 72B, Llama 3.3 70B, Qwen 2.5 VL 72B+8 more

NVIDIA A800 80GBNVIDIA upgrade

80 GB VRAM (+32)1935 GB/s (+1263)

Unlocks Mistral Small 4 119B, Qwen 2.5 72B, Llama 3.3 70B+10 more · +213% faster avg

MacBook Pro M3 Max 128GBBest value

128 GB Unified (+80)

Unlocks Mistral Small 4 119B, Qwen 2.5 72B, Llama 3.3 70B+12 more

~$2,499 MSRP

AMD Instinct MI350X 288GBBiggest leap

288 GB VRAM (+240)8000 GB/s (+7328)

Unlocks Devstral 2 123B Instruct, Mistral Small 4 119B, Qwen3-Coder 480B A35B Instruct+26 more · +1053% faster avg

~$8,000 MSRP

Compare this GPU